Srecko, I will be running a few evaluations over the next couple of months, and will post results to this list. If you have some data you'd like to include, please send it our way.
Cheers Pablo On Apr 24, 2012 8:57 PM, "Srecko Joksimovic" <[email protected]> wrote: > OK, Thank you very much! > > I will send you data that I have collected, but I need to do few more > analysis. > > Best, > Srecko > > -----Original Message----- > From: Rupert Westenthaler [mailto:[email protected]] > Sent: Tuesday, April 24, 2012 20:52 > To: Srecko Joksimovic > Cc: [email protected] > Subject: Re: precision and recall > > Hi Srecko, > > as far as I know such data are not available for Apache Stanbol. > However there might be precision/recall data for the OpenNLP models > used by Apache Stanbol. > > Over the summer there will be a lot of activity related to Entity > Disambiguation. In addition it is planed - as part of the DBpedia > Spotlight integration - to do some benchmarking of Stanbol by > using/extending the Enhancer benchmarking utility. So there might be > more information in foreseeable time. > > best > Rupert > > On Tue, Apr 24, 2012 at 8:41 PM, Srecko Joksimovic > <[email protected]> wrote: > > Hello Rupert! > > > > > > > > I would like to make few analysis, and I have defined metrics. But I > would > > like to have, if it is possible, precision and recall measures for > > annotating document corpus. If such a data are available to you, could > you > > please send it to me? It is not important what kind of document corpus > was > > analyzed. > > > > > > > > Thank you. > > > > > > > > Best, > > > > Srecko > > > > -- > | Rupert Westenthaler [email protected] > | Bodenlehenstraße 11 ++43-699-11108907 > | A-5500 Bischofshofen > >
